Experimental validation of an improved underwater sphere-target localisation scheme for an autonomous underwater vehicle manipulator system

In this study, an underwater sphere-target localisation method based on non-parallel binocular vision for an autonomous underwater vehicle manipulator system (AUVMS) is investigated. Existing extraction methods for underwater targets require considerable processing time, such that the real-time requirements for an AUVMS are not satisfied, and the shape features of the targets are not considered. In this study, an underwater target extraction method with reduced processing time is developed based on the known target edge constraint. The experimental results show that the developed method requires less time than the existing target extraction methods and satisfies the real-time requirements. In addition, the localisation precision of existing target localisation methods is relatively low, which leads to a low rate of successful implementation for an AUVMS. To solve this problem, a secondary compensation-based target localisation method is developed. A theoretical analysis is also presented. The experimental results show that the developed target localisation method can effectively improve the localisation accuracy and rate of successful implementation.